Validation of a magnetic resonance imaging-based auto-contouring software tool for gross tumour delineation in head and neck cancer radiotheraphy planning

Doshi, T. and Wilson, C. and Paterson, C. and Lamb, C. and James, A. and MacKenzie, K. and Soraghan, J. and Petropoulakis, L. and Di Caterina, G. and Grose, D. (2017) Validation of a magnetic resonance imaging-based auto-contouring software tool for gross tumour delineation in head and neck cancer radiotheraphy planning. Clinical Oncology, 29 (1). pp. 60-67. ISSN 0936-6555 (https://doi.org/10.1016/j.clon.2016.09.016)

[thumbnail of Doshi-etal-CO2016-Validation-of-a-magnetic-resonance-imaging-based-auto-contouring-software]
Preview
Text. Filename: Doshi_etal_CO2016_Validation_of_a_magnetic_resonance_imaging_based_auto_contouring_software.pdf
Accepted Author Manuscript
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo

Download (783kB)| Preview

Abstract

To perform statistical validation of a newly developed magnetic resonance imaging (MRI) auto-contouring software tool for gross tumour volume (GTV) delineation in head and neck tumours to assist in radiotherapy planning. Axial MRI baseline scans were obtained for 10 oropharyngeal and laryngeal cancer patients. GTV was present on 102 axial slices and auto-contoured using the modified fuzzy c-means clustering integrated with level set method (FCLSM). Peer reviewed (C-gold) manual contours were used as the reference standard to validate auto-contoured GTVs (C-auto) and mean manual contours (C-manual) from 2 expert clinicians (C1 and C2). Multiple geometrical metrics, including Dice Similarity Coefficient (DSC) were used for quantitative validation. A DSC ≥0.7 was deemed acceptable. Inter-and intra-variabilities amongst the manual contours were also validated. The 2-dimension (2D) contours were then reconstructed in 3D for GTV volume calculation, comparison and 3D visualisation. The mean DSC between C-gold and C-auto was 0.79. The mean DSC bet ween C-gold and C-manual was 0.79 and that between C1 and C2 was 0.80. The average time for GTV auto-contouring per patient was 8 minutes (range 6-13mins; mean 45seconds per axial slice) compared to 15 minutes (range 6-23mins; mean 88 seconds per axial slice) for C1. The average volume concordance between C-gold and C-auto volumes was 86. 51% compared to 74.16% between C-gold and C-manual. The average volume concordance between C1 and C2 volumes was 86.82%. This newly-designed MRI-based auto-contouring software tool shows initial acceptable results in GTV delineation of oropharyngeal and laryngeal tumours using FCLSM. This auto-contouring software tool may help reduce inter-and intra- variability and can assist clinical oncologists with time-consuming, complex radiotherapy planning.